The candidate's training, thus far, has involved two major themes, one as a rehabilitation scientist in the field of stroke recovery and the other as a neuroscientist with interest in revealing and modulating the underlying mechanisms. The candidate's long-term goal is to become a successful, independent clinical neuro-rehabilitation scientist specializing in customizing interventions to patient-specific mechanisms of disease and recovery. Her short term goals, proposed in this application, will further her training to help advance her towards her long term goal. As a physical therapist intrigued by the mechanisms and treatment of residual deficits of paretic upper limb in stroke, the candidate decided to pursue PhD in Rehabilitation Science with a focus in Neuroscience. She found, using functional Magnetic Resonance Imaging (fMRI), that skill learning-based rehabilitation enhanced recovery of the paretic upper limb by promoting neuroplasticity in motor cortices. In her post-doctoral fellowship, she became interested in directly modulating plasticity in the surviving cortices using non-invasive brain stimulation (Transcranial Magnetic Stimulation, TMS, and Transcranial Direct Current Stimulation, tDCS) to enhance efficacy of rehabilitation. In a small pilot study, the candidate confirmed the synergistic advantage of combining tDCS of the surviving higher visual areas with concurrent visual rehabilitation in post-stroke visual recovery. These findings engendered her interest in translating this paradigm to alleviate deficits of the paretic upper limb. The candidate proposes to apply this paradigm to improve the effectiveness of a novel, clinical rehabilitative method, called Constraint-Induced Movement Therapy (CIMT), which alleviates residual deficits by promoting adaptive plasticity in the cortical networks, but has poor clinical utility due to its labor-intensive protocols and inadequate gains. By combining cortical stimulation during CIMT, the candidate premises that its efficiency and efficacy could be improved. Unlike animal and preclinical studies, however, two clinical trials failed to support the efficacy of combining motor cortical stimulation with rehabilitation of the paretic upper limb. In a critical appraisal, the candidate concluded that discrepancies stem from over-reliance on the surviving motor cortex (M1) as a target for stimulation, which may have limited viability in humans, and lack of understanding of mechanisms of recovery. The current proposal will address these gaps in clinical research by targeting stimulation of a higher motor area, Premotor Cortex (PMC), which is remote, yet well connected to M1, has independent cortico-spinal tracts, and demonstrates adaptive plasticity with rehabilitation. Further, the candidate proposes to explore comprehensive structural and functional neural mechanisms of recovery using Diffusion Tensor Imaging (integrity of corticospinal tracts), TMS (functioning of corticospinal tracts), fMRI (balance between bilateral motor cortices), fcMRI (functional connectivity between multiple cortices) and coupling between cortical drive and paretic muscle (EEG-EMG). The unique yet complementary nature of these multi-modal techniques will increase the prediction of functional potential of recovery and reveal a complete picture of neuro-motor recovery in stroke. Thus, the research goals for the proposal are: 1) to compare the effectiveness of tDCS, delivered to surviving PMC, plus CIMT versus CIMT delivered alone in improving function of the paretic upper limb in chronic stroke. 2) To explore and contrast neural mechanisms of recovery underlying tDCS plus CIMT versus CIMT alone using multimodal structural and functional imaging from baseline to post-rehabilitation. Thirty patients will be randomly assigned to one of two groups and training will last 30 min, 3 days/week for 5 weeks. Although the candidate has gained important skills in her previous training, such as conducting laboratory-based rehabilitation studies, use of fMRI, TMS and tDCS, she requires additional training to achieve the research aims. Thus, her short-term goals are: 1) to train in Clinical Rehabilitative Research to understand clinical trial design and analyses through instruction, and research with mentors and collaborators to accomplish research aim 1 and 2) to train in Multimodal Structural and Functional Neuroimaging to learn DTI, fcMRI and EEG-EMG analyses, their interaction with fMRI and TMS, and lesions to conduct research aim 2. The candidate's short-term goals will provide her experience in conducting randomized clinical trials to create an evidence-base for neuro-rehabilitation. Further, by revealing mechanisms of recovery, an algorithm for patient-specific interventions could be devised, which will advance her towards her long-term goal. Her current position at the Cleveland Clinic provides the required support and resources for her aims. Her position as a Project Scientist offers 100% protected time for research. Collaborations across institutes, i.e., Lerner Research - Biomedical Engineering, Physical Medicine & Rehabilitation, Neurological and Imaging Institutes have created an ideal, multi-disciplinary team of established mentors and collaborators to further her learning. The mentoring team includes 1) an eminent neurophysiologist who will train the candidate in EEG-EMG, 2) a leading functional and stereotaxic neurosurgeon with clinical research expertise in applying brain stimulation who will train her to identify PMC targets in stroke brains, 3) a renowned imaging physicist with experience in developing the latest analyses for DTI and fcMRI and 4) a neuroradiologist who has developed sophisticated imaging analyses to address lesion effects.